The Effects of Extracellular Matrix on Tissue Engineering Construction of Cartilage in Vitro

In the study of bone and cartilage diseases in our laboratory, chondrocyte stereo-differentiation mod-el was cultured into cartilage tissue. The cultured cartilage tissue was closely similar to natural tissue inboth histological condition and biochemical metabolism. The model mimicked the chondrocyte differenti-ation and metabolism in vivo completely〔1〕. We studied some essential extracellular matrices, includingcollagen, proteoglycan(PG) and hyaluronic acid(HA) on the construction of grow…     

Gene expression profiling of human articular cartilage grafts generated by tissue engineering

Cartilage tissue engineering is applied clinically to cover and regenerate articular cartilage defects. In this study autologous human cartilage tissue engineering grafts based on bioresorbable polyglactin/polydioxanone scaffolds were analyzed on the broad molecular level. RNA from freshly isolated, primary and expanded adult articular chondrocytes and from three-dimensional cartilage grafts were used for gene expression profiling using oligonucleotide microarrays. The capacity of cartilage grafts to form cartilage matrix was evaluated after subcutaneous transplantation into nude mice. Gene expression profiling showed reproducibly the regulation of 905 genes and documented that chondrocytes undergo fundamental changes during cartilage tissue engineering regarding chondrocyte metabolism, growth, and differentiation. Three-dimensional assembly of expanded, dedifferentiated chondrocytes initiated the re-differentiation of cells that was accompanied by the reversal of the expression profile of multiple players of the transforming growth factor (TGF) signaling pathway including growth and differentiation factor-5 and inhibitor of differentiation-1 as well as by the induction of typical cartilage-related matrix genes such as type II collagen and cartilage oligomeric matrix protein. Cartilage grafts formed a cartilaginous matrix after transplantation into nude mice. Three-dimensional tissue culture of expanded articular chondrocytes initiates chondrocyte re-differentiation in vitro and leads to the maturation of cartilage grafts towards hyaline cartilage in vivo.     

Immunohistochemical distribution patterns of collagen type II,chondroitin 4-sulfate,laminin and fibronectin in human nasal septal cartilage

Collagen type II, chondroitin 4-sulfate, laminin and fibronectin are major components of cartilage matrix. It is important to know their distribution patterns to evaluate relationships between cartilage cells and surrounding cartilage matrix. In the present study, we investigated localization patterns of these macromolecules in human nasal septal cartilage by immunohistochemical methods. Samples of human nasal septal cartilage were obtained from patients with nasal septum deviations who underwent septoplastic operation and were prepared for immunohistochemical examination. Distribution patterns of cartilage matrix macromolecules correlated with those found in other cartilage tissues. Diffuse staining of collagen type II was found in the cartilage matrix, chondroitin 4-sulfate immunostaining was present in the cytoplasm and like a pericellular ring around chondrocytes. Laminin immunostaining was found in the cytoplasm of chondrocytes, and fibronectin was localized in the pericellular matrix and in capsules of human nasal septal cartilage. Moreover, fibronectin was also detected at high levels in the interconnecting segments between adjacent chondrons. In conclusion, similar localisation patterns of the components investigated in human septal cartilage as in other tissues indicate that these macromolecules may play a role in both cell-matrix adhesion and matrix-matrix cohesion in the pericellular microenvironment surrounding nasal septal cartilage chondrocytes as in other cartilage tissues.     

Review of injectable cartilage engineering using fibrin gel in mice and swine models

More than a decade of work has been devoted to engineering cartilage for articular surface repair. This review covers the use of fibrin gel polymer as an injectable scaffold for generating new cartilage matrix from isolated articular chondrocytes beginning with studies in mice and culminating in an applied study in swine joints. These studies began with developing a formulation of fibrin that was injectable and promoted cartilage matrix formation. Subsequent studies addressed the problems of volume loss after the scaffolds were placed in vivo by adding lyophilized cartilage matrix. Additional studies focused on the ability of isolated chondrocytes to heal and repair cartilage in a model that could be biomechanically tested. In conclusion, this series of studies demonstrated that fibrin gel is a suitable polymer gel for generating new cartilage matrix from articular chondrocytes. The new matrix is capable of forming mechanical bonds between cartilage disks and can lead to healing and integration. Armed with these results, implantation of fibrin-cell constructs into defects in swine knees showed new cartilage formation and filling of the defects. Continuing work in these models with fibrin and other polymerizable hydrogels could result in a suitable cell-based therapy for articular cartilage lesions.     

Three-dimensional tissue engineering of hyaline cartilage: comparison of adult nasal and articular chondrocytes

Adult chondrocytes are less chondrogenic than immature cells, yet it is likely that autologous cells from adult patients will be used clinically for cartilage engineering. The aim of this study was to compare the postexpansion chondrogenic potential of adult nasal and articular chondrocytes. Bovine or human chondrocytes were expanded in monolayer culture, seeded onto polyglycolic acid (PGA) scaffolds, and cultured for 40 days. Engineered cartilage constructs were processed for histological and quantitative analysis of the extracellular matrix and mRNA. Some engineered constructs were implanted in athymic mice for up to six additional weeks before analysis. Using adult bovine tissues as a cell source, nasal chondrocytes generated a matrix with significantly higher fractions of collagen type II and glycosaminoglycans as compared with articular chondrocytes. Human adult nasal chondrocytes proliferated approximately four times faster than human articular chondrocytes in monolayer culture, and had a markedly higher chondrogenic capacity, as assessed by the mRNA and protein analysis of in vitro-engineered constructs. Cartilage engineered from human nasal cells survived and grew during 6 weeks of implantation in vivo whereas articular cartilage constructs failed to survive. In conclusion, for adult patients nasal septum chondrocytes are a better cell source than articular chondrocytes for the in vitro engineering of autologous cartilage grafts. It remains to be established whether cartilage engineered from nasal cells can function effectively when implanted at an articular site.     

Role of oxygen radicals and IL-6 in IL-1-dependent cartilage matrix degradation

It has been suggested that IL-1 produces cartilage matrix degradation by metalloproteinases such as collagenase and that such degradation is regulated by metalloproteinase inhibitors. In the present study, the effects of IL-6 and oxygen radical scavengers on cartilage matrix degradation were studied. Superoxide dismutase, catalase, or methionine all significantly inhibited cartilage matrix degradation both in IL-1-stimulated and unstimulated experimental conditions. Both 10 mM EDTA and 100 nM tissue inhibitor of metalloproteinase (TIMP) significantly inhibited cartilage matrix degradation. The addition of methionine significantly inhibited collagenase activity produced in the culture supernatants of chondrocytes stimulated with IL-1. IL-6 significantly suppressed cartilage matrix degradation produced spontaneously or by IL-1 stimulation in chondrocytes. IL-6 inhibited superoxide production by chondrocytes both in IL-1-stimulated or unstimulated conditions. These results suggest that oxygen radicals are involved in cartilage matrix degradation mediated by both paracrine and autocrine IL-1 mechanisms and that oxygen radicalmediated activation of collagenase in chondrocytes may explain the mechanisms of how oxygen radicals are involved in cartilage matrix degradation. IL-6 inhibited superoxide production in chondrocytes and thus inhibited cartilage matrix degradation.     

Morphological modifications during long-term survival of Meckel's cartilage hypertrophic chondrocytes transplanted in the mouse spleen.

To examine the ultimate fate of hypertrophic chondrocytes, Meckel's cartilage bars from 18-day-old mouse embryos were transplanted into isogenic mouse spleen for 3, 7, 14 and 21 days and observed at the light and electron microscopic levels. The midportions of these Meckel's cartilage bars were used as explants; they were characterized by many hypertrophic chondrocytes containing euchromatic round nuclei, a large amount of glycogen particles, and some vacuoles. Grafted cartilage adapted well to the splenic tissue, showing intense metachromasia around the territorial matrix. Ultrastructural observations indicated that the number of large vacuoles and glycogen aggregates in the hypertrophic cells became markedly reduced with grafting time, whereas the Golgi apparatus and rough endoplasmic reticulum were well-developed. Needle-like crystals showing initial apatite deposition appeared in association with matrix vesicles; these proliferated as time elapsed after transplantation. On day 14 after transplantation, cells displaying such various structural features as pyknotic nuclei, large vacuoles, and cytoplasmic shrinkage were noted in addition to intact hypertrophic chondrocytes. Following resorption of the calcified cartilage by multinucleated giant cells, many osteoblasts appeared along the border of the calcified matrix. Some remaining hypertrophic cells in the calcified matrix had transformed into osteocyte-like cells. On day 21, the resorbed area of the calcified cartilage was invaded by many blood vessels. Hypertrophic chondrocytes, now exposed from cellular lacunae, and the osteocyte-like cells in the calcified matrix displayed involutional changes. The present study showed that, although the hypertrophic chondrocytes in Meckel's cartilage essentially underwent regressive changes, they retained the ability to stimulate endochondral ossification within the microenvironment of the spleen. In addition, some of these cells were transformed into osteocyte-like cells.     

Human platelet supernatant promotes proliferation but not differentiation of articular chondrocytes

The objective of the study was to evaluate the growth-promoting activity of human platelet supernanant on primary chondrocytes in comparison with fetal calf serum (FCS) supplemented cell culture medium. Furthermore, the differentiation potential of platelet supernatant was determined in three-dimensional artificial cartilage tissues of bovine articular chondrocytes. Proliferation of articular and nasal septal chondrocytes was assayed by incorporation of BrdU upon stimulation with ten different batches of human platelet supernatant. On bovine articular chondrocytes, all these batches were at least as growth-promoting as FCS. On nasal septal chondrocytes, nine out of ten batches revealed increased or equivalent mitogenic stimulation compared with medium supplemented with FCS. Three-dimensional culture and subsequent histological analysis of matrix formation were used to determine the differentiation properties of platelet supernatant on articular chondrocytes. Human platelet supernatant failed to induce the deposition of typical cartilage matrix components, whereas differentiation and matrix formation were apparent upon cultivation of articular chondrocytes with FCS. Proliferation assays demonstrated that human platelet supernatant stimulates growth of articular and nasal septal chondrocytes; however, platelet supernatant failed to stimulate articular chondrocytes to redifferentiate in three-dimensional chondrocyte cultures. Therefore platelet lysate may be suitable for chondrocyte expansion, but not for maturation of tissue-engineered cartilage.     

Localization of Mycoplasma pulmonis in cartilage.

Transmission electron microscopy of articular cartilage in Mycoplasma pulmonis-infected neonatal rats revealed the presence of mycoplasmas within the matrix and lacunae. The mycoplasmas appeared to have a tropism for the chondrocytes and induced lysis of both the chondrocytes and matrix of the cartilage.     

Behavior of human articular chondrocytes derived from nonarthritic and osteoarthritic cartilage in a collagen matrix

The purpose of this study was to evaluate the morphologic and biochemical behavior and activity of human chondrocytes taken from nonarthritic and osteoarthritic cartilage and seeded on a three-dimensional matrix consisting of collagen types I, II, and III. Human articular chondrocytes were isolated from either nonarthritic or osteoarthritic cartilage of elderly subjects, and from nonarthritic cartilage of an adolescent subject, seeded on collagen matrices, and cultured for 12 h, 7 days, and 14 days. Histological analysis, immunohistochemistry, and biochemical assays for glycosaminoglycans (GAGs) and DNA content were performed for cell-seeded and unseeded matrices. Chondrocytes of nonarthritic cartilage revealed a larger number of spherical cells, consistent with a chondrocytic phenotype. The biochemical assay showed a net increase in GAG content in nonarthritic chondrocytes, whereas almost no GAGs were seen in osteoarthritic cells. The DNA results suggest that more osteoarthritic cells than chondrocytes from nonarthritic cartilage attached to the matrix within the first week. Human articular chondrocytes isolated from osteoarthritic cartilage seem to have less bioactivity after expansion and culture in a sponge consisting of type I, II, and III collagen compared with chondrocytes from nonarthritic cartilage.     

Localization of type I and II collagen during development of human first rib cartilage

The localization of fibrillar type I and II collagen was investigated by immunofluorescence staining with specific antibodies in order to obtain a better understanding of tissue remodelling during the development of first rib cartilage. In childhood and early adolescence type I collagen was found to be restricted to the perichondrium of first rib cartilage, while type II collagen was localized in the matrix of hyaline cartilage. However, in advanced age type I collagen was also found in the territorial matrix of intermediate and central chondrocytes of first rib cartilage. The matrix of subperichondrial chondrocytes was negative for type I collagen. This suggests that some chondrocytes in first rib cartilage undergo a modulation to type I collagen-producing cells. The first bone formation was observed in rib cartilages of 20- to 25-year-old adults. Interestingly, the ossification began peripherally, adjacent to the innermost layer of the perichondrium where areas of fibrocartilage had developed. The newly formed bone matrix showed strong immunostaining for type I collagen. Fibrocartilage bordering peripherally on bone matrix revealed only a faint staining for type I collagen, but strong immunoreactivity to type II collagen. The interterritorial matrix of the central chondrocytes failed to react with the type II collagen antibody, in both men and women, from the end of the second decade. These observations indicate that major matrix changes occur at the same time in male and female first rib cartilages. Thus, our findings indicate that ossification in human first rib cartilage does not follow the same pattern as that observed in endochondral ossification of epiphyseal discs or sternal cartilage.     

Biomechanical analysis of a chondrocyte-based repair model of articular cartilage.

The objective of this study was to evaluate the biomechanical properties of newly formed cartilaginous tissue synthesized from isolated chondrocytes. Cartilage from articular joints of lambs was either digested in collagenase to isolated chondrocytes or cut into discs that were devitalized by multiple freeze-thaw cycles. Isolated cells were incubated in suspension culture in the presence of devitalized cartilage matrix for 3 weeks. Multiple chondrocyte/matrix constructs were assembled with fibrin glue and implanted subcutaneously in nude mice for up to 6 weeks. Testing methods were devised to quantify integration of cartilage pieces and mechanical properties of constructs. These studies showed monotonic increase with time in tensile strength, fracture strain, fracture energy, and tensile modulus to values 5-10% of normal articular cartilage by 6 weeks in vivo. Histological analysis indicated that chondrocytes grown on dead cartilage matrix produced new matrix that integrated individual cartilage pieces with mechanically functional tissue.     

Integrative repair of cartilage with articular and nonarticular chondrocytes

Articular chondrocytes can synthesize new cartilaginous matrix in vivo that forms functional bonds with native cartilage. Other sources of chondrocytes may have a similar ability to form new cartilage with healing capacity. This study evaluates the ability of various chondrocyte sources to produce new cartilaginous matrix in vivo and to form functional bonds with native cartilage. Disks of articular cartilage and articular, auricular, and costal chondrocytes were harvested from swine. Articular, auricular, or costal chondrocytes suspended in fibrin glue (experimental), or fibrin glue alone (control), were placed between disks of articular cartilage, forming trilayer constructs, and implanted subcutaneously into nude mice for 6 and 12 weeks. Specimens were evaluated for neocartilage production and integration into native cartilage with histological and biomechanical analysis. New matrix was formed in all experimental samples, consisting mostly of neocartilage integrating with the cartilage disks. Control samples developed fibrous tissue without evidence of neocartilage. Ultimate tensile strength values for experimental samples were significantly increased (p < 0.05) from 6 to 12 weeks, and at 12 weeks they were significantly greater (p < 0.05) than those of controls. We conclude that articular, auricular, and costal chondrocytes have a similar ability to produce new cartilaginous matrix in vivo that forms mechanically functional bonds with native cartilage.     

The influence of bone and marrow on cartilage hypertrophy and degradation during 30-day serum-free culture of the embryonic chick tibia.

In this study, an organ culture system is defined which demonstrates complete loss of cartilage matrix from embryonic chick tibiae. Efficient loss of the cartilage matrix occurs within 30 days of serum-free culture only when the intact tibiae containing bone, marrow, and cartilage tissue are cultured. During organ culture nonhypertrophic chondrocytes become hypertrophic and stain positively for type X collagen and alkaline phosphatase. The cartilage loses Safranin O staining, and finally all cartilage matrix disappears leaving the bony collar and marrow cells. If the tibial cartilage is separated from the bony collar and cultured alone in serum-free medium, the nonhypertrophic chondrocytes also hypertrophy; the matrix loses Safranin O staining; however, some components of the matrix including type X collagen still remain after 30 days. In the presence of serum, the chondrocytes will hypertrophy but cartilage degradation is not evident. The results of this study support the conclusions that 1) hypertrophy is inherently programmed in the chondrocyte and 2) while Safranin O staining of cartilage cultured alone is diminished in serum-free organ culture, the degradation of cartilage is complete only when bone and marrow are also present.     

Berakdown of articular cartilage by vascular tissue

A study has been made with porcine tissues in organ culture of th effect of vascular tissue on articular cartilage. Isolated explants of cartilage were maintained alive for 14 days without breakdown of the matrix. Contact with vascular tissue produced extensive loss of proteoglycan and collagen wth fibroblastic transformaton of te chondrocytes. The breakdown was partly dur to a direct effect on te matrix and partly to activation of catabolic processes in the chondrocytes. Vascular tissue produced exactly the same loss of matrix as synovial tissue.     

Cell-based tissue-engineered allogeneic implant for cartilage repair

The potential for using of allogeneic cartilage chips, transplanted in a biologic polymer with articular chondrocytes, as a tool for articular cartilage repair was studied. Small lyophilized articular cartilage chips were mixed with a cell/fibrinogen solution and thrombin to obtain implantable constructs made of fibrin glue, chondrocytes, and cartilage chips. Specimens were implanted in the subcutaneous tissue on the backs of nude mice (experimental group A). Three groups of controls (groups B, C, and D) were also prepared. Group B consisted of fibrin glue and cartilage chips without chondrocytes. Group C consisted of fibrin glue and chondrocytes without cartilage chips, and group D was composed solely of fibrin glue. All samples were carefully weighed before implantation in the mice. The constructs were harvested from the animals at 6, 9, and 12 weeks, examined grossly, and weighed. The samples were then processed and stained with hematoxylin and eosin for histological examination. Gross evaluation and weight analysis of the constructs at the time of retrieval showed retention of the original mass in the samples made of fibrin glue, chondrocytes, and cartilage chips (group A) and demonstrated a cartilaginous consistency upon probing. Specimens from constructs of fibrin glue and cartilage chips without chondrocytes (control group B) retained most of their volume, but were statistically lighter than specimens from group A and were much softer and more pliable than those in group A. Samples of specimens from constructs of fibrin glue and chondrocytes (groups C) and fibrin glue alone (group D) both showed a substantial reduction of their original masses over the experimental time periods when compared to the samples in groups A and B, although specimens from group C demonstrated new cartilage matrix formation. Histological analysis of specimens in experimental group A demonstrated the presence of cartilage chips surrounded by newly formed cartilaginous matrix, while specimens of control group B showed only fibrotic tissue surrounding the devitalized cartilage pieces. Cartilaginous matrix was also observed in control group C, in which cartilage chips were absent, whereas only fibrin glue debris was observed in control group D. This study demonstrated that a composite of fibrin glue and devitalized cartilage can serve as a scaffold for chondrocyte transplantation, preserve the original phenotype of the chondrocytes, and maintain the original mass of the implant. This may represent a valid option for addressing the problem of articular cartilage repair.     

Structural colocalisation of type VI collagen and fibronectin in agarose cultured chondrocytes and isolated chondrons extracted from adult canine tibial cartilage

Cell–matrix and matrix–matrix interactions are of critical importance in regulating the development, maintenance and repair of articular cartilage. In this study, we examined the structural colocalisation of type VI collagen and fibronectin in isolated chondrons and long-term agarose cultured chondrocytes extracted from normal adult canine articular cartilage. Using double labelling immunohistochemistry in conjunction with dual channel confocal microscopy and digital image processing we demonstrate that type VI collagen and fibronectin are distributed in a similar staining pattern and are colocalised at the surface of cultured chondrocytes and isolated chondrons. The results suggest that type VI collagen and fibronectin may play a role in both cell–matrix adhesion and matrix–matrix cohesion in the pericellular microenvironment surrounding articular cartilage chondrocytes.     

Differentiation capacity of human chondrocytes embedded in alginate matrix

Healing capacity of cartilage is low. Thus, cartilage defects do not regenerate as hyaline but mostly as fibrous cartilage which is a major drawback since this tissue is not well adapted to the mechanical loading within the joint. During in vitro cultivation in monolayers, chondrocytes proliferate and de-differentiate to fibroblasts. In three-dimensional cell cultures, de-differentiated chondrocytes could re-differentiate toward the chondrogenic lineage and re-express the chondrogenic phenotype. The objective of this study was to characterize the mesenchymal stem cell (MSC) potential of human chondrocytes isolated from articular cartilage. Furthermore, the differentiation capacity of human chondrocytes in three-dimensional cell cultures was analyzed to target differentiation direction into hyaline cartilage. After isolation and cultivation of chondrogenic cells, the expression of the MSC-associated markers: cluster of differentiation (CD)166, CD44, CD105, and CD29 was performed by flow cytometry. The differentiation capacity of human chondrocytes was analyzed in alginate matrix cultured in Dulbecco?s modified eagle medium with (chondrogenic stimulation) and without (control) chondrogenic growth factors. Additionally, the expression of collagen type II, aggrecan, and glycosaminoglycans was determined. Cultivated chondrocytes showed an enhanced expression of the MSC-associated markers with increasing passages. After chondrogenic stimulation in alginate matrix, the chondrocytes revealed a significant increase of cell number compared with unstimulated cells. Further, a higher synthesis rate of glycosaminoglycans and a positive collagen type II and aggrecan immunostaining was detected in stimulated alginate beads. Human chondrocytes showed plasticity whilst cells were encapsulated in alginate and stimulated by growth factors. Stimulated cells demonstrated characteristics of chondrogenic re-differentiation due to collagen type II and aggrecan synthesis.     

Temporal and spatial localization of type I and II collagens in human thyroid cartilage

Thyroid cartilages of various ages were investigated by immunofluorescence staining for localization of the fibrillar collagen types I and II in order to understand the tissue remodeling occurring during the mineralization and ossification of thyroid cartilage. In fetal and juvenile thyroid cartilages, type I collagen was restricted to the inner and outer perichondrium, while type II collagen was localized in the matrix of hyaline cartilage. However, in advanced ages, type I collagen was also localized in the pericellular and in the interterritorial matrix of intermediate and central chondrocytes of thyroid cartilage. The matrix of peripheral chondrocytes was negative for type I collagen. This suggests that some chondrocytes in thyroid cartilage undergo a differentiation to type I collagen-producing chondrocytes. At the beginning of ossification, bone-related type I collagen was chiefly detected in the central cartilage layer, but was never deposited first from the perichondrium in the direction to the subperichondrial cartilage. This observation confirmed previous findings showing that osteogenesis mainly follows an endochondral ossification pattern. Interterritorial matrix failed to react with the type II collagen antibody in men from the beginning of the third decade, and later still in women, even after treatment with hyaluronidase. These observations indicate that major matrix changes occur faster in male than in female thyroid cartilage.Dedicated to Professor Dr. W. Kühnel on the occasion of his 60th birthday